AUGMENTATION AND CLASSIFICATION OF TIME SERIES FOR FINDING ACL INJURIES
2022 (Engelska)Självständigt arbete på grundnivå (kandidatexamen), 10 poäng / 15 hp
Studentuppsats (Examensarbete)
Abstract [en]
This thesis addresses the problem where we want to apply machine learning over a small data set of multivariate time series. A challenge when classifying data is when the data set is small and overfitting is at risk. Augmentation of small data sets might avoid overfitting. The multivariate time series used in this project represent motion data of people with reconstructed ACLs and a control group. The approach was pairing motion data from the training set and using Euclidean Barycentric Averaging to create a new set of synthetic motion data so as to increase the size of the training set. The classifiers used were Dynamic Time Warping -One Nearest neighbour and Time Series Forest. In our example we found this way of increasing the training set a less productive strategy. We also found Time Series Forest to generally perform with higher accuracy on the chosen data sets, but there may be more effective augmentation strategies to avoid overfitting.
Ort, förlag, år, upplaga, sidor
2022. , s. 32
Serie
UMNAD ; 1330
Nyckelord [en]
computer science, machine learning, motion analysis, reconstructed ACL, anterior cruciate ligament, time series forest, dynamic time wapring, ACL, multivariate time series clasification, MTSC, time series classification, TSC, euclidean barycentric average, euclidean barycentric averaging, autmentation of time series, augmentation of multivariate time series, data augmentation, augmentation
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
URN: urn:nbn:se:umu:diva-197105OAI: oai:DiVA.org:umu-197105DiVA, id: diva2:1674949
Externt samarbete
FoU Region Västerbotten
Utbildningsprogram
Kandidatprogrammet i Datavetenskap
Handledare
Examinatorer
2022-06-232022-06-222022-06-23Bibliografiskt granskad